ICML

MoHAVE: Mixture of Hierarchical Audio-Visual Experts for Robust Speech Recognition

Sungnyun Kim, Kangwook Jang, Sangmin Bae, Sungwoo Cho, Se-Young Yun

Comments Accepted to ICML 2025

PlaySlot: Learning Inverse Latent Dynamics for Controllable Object-Centric Video Prediction and Planning

Angel Villar-Corrales, Sven Behnke

机构 * Autonomous Intelligent Systems, Computer Science Institute VI – Intelligent Systems（智能系统） ； Robotics, Center for Robotics（机器人学） ； the Lamarr Institute for Machine Learning（拉马尔机器学习研究所） ； Artificial Intelligence, University of Bonn, Germany（人工智能，波恩大学，德国） 纠错

Comments ICML 2025

Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding

Mingyu Jin, Kai Mei, Wujiang Xu, Mingjie Sun, Ruixiang Tang, Mengnan Du, Zirui Liu, Yongfeng Zhang

Comments International Conference on Machine Learning (ICML 2025)

How Contaminated Is Your Benchmark? Quantifying Dataset Leakage in Large Language Models with Kernel Divergence

Hyeong Kyu Choi, Maxim Khanov, Hongxin Wei, Yixuan Li

Comments ICML 2025

Ca2-VDM: Efficient Autoregressive Video Diffusion Model with Causal Generation and Cache Sharing

Kaifeng Gao, Jiaxin Shi, Hanwang Zhang, Chunping Wang, Jun Xiao, Long Chen

Comments Accepted by ICML 2025. Code is available: https://github.com/Dawn-LX/CausalCache-VDM

Beyond The Rainbow: High Performance Deep Reinforcement Learning on a Desktop PC

Tyler Clark, Mark Towers, Christine Evers, Jonathon Hare

Comments 9 main pages, 28 total. Accepted at ICML 2025 (Poster)

EvoMesh: Adaptive Physical Simulation with Hierarchical Graph Evolutions

Huayu Deng, Xiangming Zhu, Yunbo Wang, Xiaokang Yang

机构 * Huayu Deng（邓华宇） ； Xiangming Zhu（朱向明） ； Yunbo Wang（王云波） ； Xiaokang Yang（杨晓康） 纠错

Comments Published as a conference paper at ICML 2025

Optimizing Adaptive Attacks against Watermarks for Language Models

Abdulrahman Diaa, Toluwani Aremu, Nils Lukas

Comments To appear at the International Conference on Machine Learning (ICML'25)

Quantum Optimization via Gradient-Based Hamiltonian Descent

Jiaqi Leng, Bin Shi

Comments 19 pages, 6 figures. To appear in the proceedings of ICML 2025

Layer-wise Quantization for Quantized Optimistic Dual Averaging

Anh Duc Nguyen, Ilia Markov, Frank Zhengqing Wu, Ali Ramezani-Kebrya, Kimon Antonakopoulos, Dan Alistarh, Volkan Cevher

Comments Accepted at the International Conference on Machine Learning (ICML 2025)

Enhancing Classification with Semi-Supervised Deep Learning Using Distance-Based Sample Weights

Aydin Abedinia, Shima Tabakhi, Vahid Seydi

机构 * Dept. of Computer Science Islamic Azad University South-Tehran Branch（计算机科学系，伊斯兰教 Azad 大学南德黑兰分校） ； School of Electrical, Computer and Biomedical Engineering Southern Illinois University（电气、计算机和生物医学工程学院，南方伊利诺伊大学） ； School of Ocean Sciences Bangor University（海洋科学学院，Bangor 大学） 纠错

Comments 5 pages, 6 figures. This paper has been accepted for publication and oral presentation at the 2025 10th IEEE International Conference on Machine Learning Technologies (ICMLT 2025). The final authenticated version will be available in IEEE Xplore following the conference

When Do LLMs Help With Node Classification? A Comprehensive Analysis

Xixi Wu, Yifei Shen, Fangzhou Ge, Caihua Shan, Yizhu Jiao, Xiangguo Sun, Hong Cheng

Comments Accepted by ICML 2025

Sequential Kernelized Independence Testing

Aleksandr Podkopaev, Patrick Blöbaum, Shiva Prasad Kasiviswanathan, Aaditya Ramdas

机构 * Carnegie Mellon University（卡内基梅隆大学） ； Amazon Web Services（亚马逊网络服务） 纠错

Comments ICML 2023

Rethink the Role of Deep Learning towards Large-scale Quantum Systems

Yusheng Zhao, Chi Zhang, Yuxuan Du

Comments ICML 2025

Guided Search Strategies in Non-Serializable Environments with Applications to Software Engineering Agents

Karina Zainullina, Alexander Golubev, Maria Trofimova, Sergei Polezhaev, Ibragim Badertdinov, Daria Litvintseva, Simon Karasik, Filipp Fisin, Sergei Skvortsov, Maksim Nekrashevich, Anton Shevtsov, Boris Yangel

Comments ICML

FreeMesh: Boosting Mesh Generation with Coordinates Merging

Jian Liu, Haohan Weng, Biwen Lei, Xianghui Yang, Zibo Zhao, Zhuo Chen, Song Guo, Tao Han, Chunchao Guo

Comments Accepted by ICML 2025, camera-ready version

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation

Jingyu Liu, Beidi Chen, Ce Zhang

Comments Proceedings of the 42nd International Conference on Machine Learning (ICML 2025)

AutoAL: Automated Active Learning with Differentiable Query Strategy Search

Yifeng Wang, Xueying Zhan, Siyu Huang

机构 * Yifeng Wang ； Xueying Zhan ； Siyu Huang 纠错

Comments ICML 2025

RATE: Causal Explainability of Reward Models with Imperfect Counterfactuals

David Reber, Sean Richardson, Todd Nief, Cristina Garbacea, Victor Veitch

Comments ICML 2025. Code at https://github.com/toddnief/RATE

Learning Utilities from Demonstrations in Markov Decision Processes

Filippo Lazzati, Alberto Maria Metelli

Comments International Conference on Machine Learning 42 (ICML 2025)

From Words to Worlds: Compositionality for Cognitive Architectures

Ruchira Dhar, Anders Søgaard

Comments Accepted to ICML 2024 Workshop on LLMs & Cognition

Temporal Distance-aware Transition Augmentation for Offline Model-based Reinforcement Learning

Dongsu Lee, Minhae Kwon

机构 * Carnegie Mellon University, Pittsburgh, USA（卡内基梅隆大学） ； Department of Intelligent Semiconductors, Soongsil University, Seoul, South Korea（智能半导体系，顺斯大学） 纠错

Comments 2025 ICML

Does Low Rank Adaptation Lead to Lower Robustness against Training-Time Attacks?

Zi Liang, Haibo Hu, Qingqing Ye, Yaxin Xiao, Ronghua Li

机构 * Zi Liang（张利亚） ； Haibo Hu（胡浩波） ； Qingqing Ye（叶青青） ； Yaxin Xiao（肖亚欣） ； Ronghua Li（李荣华） 纠错

Comments To appear at ICML 25

Deep Unfolding with Kernel-based Quantization in MIMO Detection

Zeyi Ren, Jingreng Lei, Yichen Jin, Ermo Hua, Qingfeng Lin, Chen Zhang, Bowen Zhou, Yik-Chung Wu

Comments submitted to ICML ML4Wireless workshop

High Dynamic Range Novel View Synthesis with Single Exposure

Kaixuan Zhang, Hu Wang, Minxian Li, Mingwu Ren, Mao Ye, Xiatian Zhu

Comments It has been accepted by ICML 2025

Provably Near-Optimal Federated Ensemble Distillation with Negligible Overhead

Won-Jun Jang, Hyeon-Seo Park, Si-Hyeon Lee

Comments Accepted (Poster) at ICML 2025

Function Encoders: A Principled Approach to Transfer Learning in Hilbert Spaces

Tyler Ingebrand, Adam J. Thorpe, Ufuk Topcu

Comments Final submission ICML 2025

Origin Identification for Text-Guided Image-to-Image Diffusion Models

Wenhao Wang, Yifan Sun, Zongxin Yang, Zhentao Tan, Zhengdong Hu, Yi Yang

机构 * University of Technology Sydney（悉尼技术大学） ； Peking University（北京大学） ； Harvard University（哈佛大学） ； Zhejiang University（浙江大学） 纠错

Comments Accepted by ICML 2025

Brittle Minds, Fixable Activations: Understanding Belief Representations in Language Models

Matteo Bortoletto, Constantin Ruhdorfer, Lei Shi, Andreas Bulling

机构 * University of Stuttgart（斯图加特大学） 纠错

Comments ICML 2024 Workshop on Mechanistic Interpretability version: https://openreview.net/forum?id=yEwEVoH9Be

Action-Dependent Optimality-Preserving Reward Shaping

Grant C. Forbes, Jianxun Wang, Leonardo Villalobos-Arias, Arnav Jhala, David L. Roberts

Comments Extended abstract at AAMAS 2025; full paper at ICML 2025